feat: add AWS Bedrock provider for foundational model integration#6122
Closed
lalalic wants to merge 3 commits intokeephq:mainfrom
Closed
feat: add AWS Bedrock provider for foundational model integration#6122lalalic wants to merge 3 commits intokeephq:mainfrom
lalalic wants to merge 3 commits intokeephq:mainfrom
Conversation
…ephq#5190) Adds a new AWS Bedrock provider that enables Keep workflows to query AWS foundational models (Claude, Llama, Mistral, Cohere, Titan). ## Features - Supports all major Bedrock model families with model-specific request/response handling - Auth via explicit AWS credentials or IAM role/instance profile (zero-config for EC2/ECS/EKS) - Optional session token support for temporary credentials - Structured output (JSON schema) support for Claude models - Lazy boto3 client initialization with caching ## Supported Models | Family | Example Model IDs | |--------|------------------| | Anthropic Claude | anthropic.claude-3-sonnet-20240229-v1:0, claude-3-haiku, claude-3-opus | | Meta Llama | meta.llama3-8b-instruct-v1:0, llama3-70b | | Mistral | mistral.mistral-7b-instruct-v0:2, mixtral-8x7b | | Cohere Command | cohere.command-r-v1:0, command-r-plus | | Amazon Titan | amazon.titan-text-express-v1, titan-text-lite | ## Tests 27 unit tests covering auth config, client creation, request body building, response parsing, and query execution for all model families. Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
|
@lalalic is attempting to deploy a commit to the KeepHQ Team on Vercel. A member of the Team first needs to authorize it. |
dosubot
Bot
added
size:XL
- Add bedrock-provider.mdx with setup guide (IAM role, explicit credentials, temporary credentials), supported model table, and workflow examples - Add auto-generated snippet (authentication params, workflow step) - Add bedrock-provider to mint.json navigation
Contributor
Author
|
Added the missing pieces:
The documentation was already added in the previous commit:
All docs validation passes. Ready for re-review! |
Member
|
Closing: AI-generated spam. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds a new AWS Bedrock provider that enables Keep workflows to query AWS foundational models, Resolves #5190.
AWS Bedrock lets teams use Claude, Llama, Mistral, Cohere, and Titan models via their own AWS account — without routing through third-party APIs. This is valuable for teams with strict data residency requirements or existing AWS infrastructure.
Supported Model Families
anthropic.claude-3-sonnet-20240229-v1:0,claude-3-haiku,claude-3-opusmeta.llama3-8b-instruct-v1:0,llama3-70bmistral.mistral-7b-instruct-v0:2,mixtral-8x7bcohere.command-r-v1:0,command-r-plusamazon.titan-text-express-v1,titan-text-liteFeatures
Auth Configuration
regionus-east-1)access_keysecret_access_keysession_tokenTests
27 unit tests with full mocking — no AWS credentials required to run.
Covers: auth config validation, boto3 client creation (credentials + IAM), request body building for all 5 model families, response parsing for all families, structured output, error handling (ClientError → ProviderException), and client caching.